3.6 and 3.8 to be release soon ?

[juin]

I found 3 retailer that annouce P4 570 (3.8)and 560 (3.6).I guess intel will annouce something in the coming week.
i need to change useur name.<P ID="edit"><FONT SIZE=-1><EM>Edited by juin on 05/20/04 08:56 AM.</EM></FONT></P>

 

[Mr_Nuke]

Cool!!!

 

[endyen]

le cinquieme de juin

 

[Mr_Nuke]

June the 5th??

 

[juin]

lien SVP

i need to change useur name.

 

[HardWareBoss]

Like way kool mon.

 

[Mephistopheles]

Wouldn't it make more sense for those releases to accompany the Alderwood/Grantsdale combo? I mean, it would give them an extra punch if they managed to launch like a 3.46 or 3.6 or whatever prescott with grantsdale. Maybe even a 3.73. This is the kind of punch they actually need to compete agressively against AMD anyway...

<i><font color=red>You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete</font color=red> - Buckminster Fuller </i>

 

[juin]

Actualy they are all Socket 775 those i have see june is in 3 week.

i need to change useur name.

 

[Mephistopheles]

Um, it would be good for Intel if they managed to get the 3.8Ghz P4 out the door by June... It might match A64's capabilities reasonably.

They still need to enable 64-bit support for marketing reasons, though.

Also, it might be a good idea for them come out not only with a 3.8Ghz prescott now, but with the 1066Mhz FSB one, the 3.73Ghz variant, if that is at all possible... Just to look as if they're trying to improve themselves... They need to get some momentum.

Wasn't Nocona supposed to be launched at 3.6Ghz? Thus it would be very unusual if there wasn't at least a 3.6Ghz P4 by the time Nocona gets on its feet.....

<i><font color=red>You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete</font color=red> - Buckminster Fuller </i><P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 05/20/04 01:55 PM.</EM></FONT></P>

 

[trooper11]

lets realise something here. if thats true, you still wont be able to buy one, if the trend sticks with the 3.2 and 3.4 p4e's. itll be weeks or even a month or two before its in any quantity. i guess its all intel needs though, just a paper saying they are better agian until they can actaully get good parts out, just keep the pot mixing with uncertainty lol.

 

[mrmonsoon]

I have a few questions for ya. I thought Intel stopped two next gen cpu's becouse of heat issues. We all know the heat issues Presscot has-won't pushing clocks higher create more heat? If this keeps up, in a year or two all new cpu's will require water cooling just to run.
Here is an idea-Intel-work on making your cpus run cooler and don't just jack up the Mhz speed, make each clock cycle do more work-like AMD.

 

[Mephistopheles]

That's why they announced the dual-core CPUs for ASAP. These will probably be some sort of Dothan-derivatives with enhancements like 64-bit extensions and very much improved memory subsystems.

In terms of heat dissipation, just think that a quad-core Dothan with 8MB cache (4x2MB) would dissipate around 86W, which is still less than any top-end CPU nowadays. If Intel considered dual-core Dothans, that would be only 43W, and they might just be able to crank up the clock to make this processor a nicely powerful one... Assuming heat is what's avoiding the clock increases.... They better be more careful now, 'cause they can't afford to have to do yet another U-turn in design ideology.

<i><font color=red>You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete</font color=red> - Buckminster Fuller </i>

 

[ZER0]

actually at an intel training event i attended last year they said they would be going to water cooling within a few years. they said they had a cheep solution that didn't use much water at all. only time will tell i guess.

 

[Mephistopheles]

It is much smarter to use better cooling solutions if they're as convenient and as cheap as the usual fans, of course. So it wouldn't hurt to develop more sophisticated cooling solutions!

...but, of course, this doesn't mean that the core engineers may stop worrying about power dissipation at all. 150W is a very, very big amount of heat.... on a small 1cm^2 surface...

<i><font color=red>You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete</font color=red> - Buckminster Fuller </i>

 

[americium]

AFAIK June 21st will be the day Intel anounce the P4 3.6GHz in LGA775 socket.

at August will be the anouncement of P4 3.8GHz LGA775
at October will be the anouncement of P4 4.0GHz LGA775

so i don't know where you seen that P4 3.8GHz will be anounced at June 21st but it sounds completely unlikely since Intel (almost) never anounced a product before its schedualed date.

yet again, it's Intel! and it's just a paper launch. so who knows. we'll have to wait for June 21st in order to know.



<P ID="edit"><FONT SIZE=-1><EM>Edited by americium on 05/20/04 04:37 PM.</EM></FONT></P>

 

[juin]

Actualie intel have say P8 will be release in 2006.P7 or netburst or commercial name Pentium 4 will be replace for a larger SMT impletation and dual core.Nephalem architecture is built from bottom to top to soly use TLP.Personnaly i wont be surprise if nephalem is a single core with cluster Int units and a large 4 Flop\cycle FPU (same as Itanium mackinley and higher).

MGHZ is very important vector instruction or data paralism are not super scalar it mean only instruction can be done at the time so clock speed is by far about 80% only way to improve it performance under those code (game rendering video audio).


Heat probleme are prescott only.Yes a good part come from the fact that Dielectrique leak a lot.About 5 to 6 layer of atoms is just getting too small.The new stepping offer better performance.Tejas was cancel for many reason heat too much logic transistor and others probleme.Also they dont have cancel any generation just a core revision.

i need to change useur name.